How Does Modification Affect the Processing of Formulaic Language? Evidence from L1 and L2 Speakers of Chinese

It has been well documented that formulaic language (FL) – strings of language above the word level in most cases, such as binomials, collocations, n-grams, idioms – enjoys a processing advantage over novel language (e.g., Arnon & Snider, 2010; Carrol & Conklin, 2020; Hallin & Van Lancker Sidtis, 2017; Siyanova-Chanturia, Conklin, & Schmitt, 2011a; Siyanova-Chanturia, Conklin, & van Heuven, 2011b; Tremblay, Derwing, Libben, & Westbury, 2011). The majority of these studies focused on formulaic sequences (FSs) in their original form. In natural language use, however, many FSs are modified with words intervening in-between the individual constituents (e.g., provide information → provide some of the information, see Vilkaitė, 2016a). Whether or not the processing advantage can be extended to modified FSs remains poorly investigated. In this thesis, three reading studies were conducted to address this gap from the perspective of Chinese FL processing, in which the influence of phrase frequency, modification degree, age, and L2 proficiency, were explored and discussed in detail.

Study 1 recorded the eye movements of L1 Chinese adults when reading sentences embedded with frequent collocations and infrequent controls. The results suggested a significant processing advantage for collocations over relative controls. Critically, the processing advantage for collocations in their original form extended to their short- (with two Chinese characters inserted in the middle of the phrase) and long-insertion forms (with four Chinese characters inserted in the middle of the phrase). The processing advantage was largely observed in the whole phrase in the late processing stage.

Study 2 and Study 3 borrowed the materials from Study 1 and applied them in a self-paced reading experiment with L2 Chinese learners from Japan and Thailand of different language proficiencies, and L1 Chinese children at Grades Three and Six, respectively. The results of Study 2 revealed a significant processing advantage for collocations over respective controls in L2 learners. More crucially, the processing advantage for L2 collocations persisted in their short- and long-insertion forms. L2 proficiency and L1 background were also found to play a role in L2 FS processing.

The results of Study 3 suggested a significant processing advantage for collocations over respective controls in L1 children. More importantly, the processing advantage for collocations remained in their short- and long-insertion forms in sixth graders; and remained in their short-insertion form in third graders. Furthermore, age was found to play a role in FS processing in children.

Contributions to the current literature and methodologies are discussed at the end of the thesis.